LAY THEORIES CONCERNING CAUSES AND TREATMENT OF DEPRESSION

Transcription

1 Journal of Rational-Emotive & Cognitive-Behavior Therapy Volume 17, Number 4, Winter 1999 LAY THEORIES CONCERNING CAUSES AND TREATMENT OF DEPRESSION Lindsey Kirk Cindy Brody Ari Solomon David A. F. Haaga American University ABSTRACT: This study compared perceptions of the causes of, therapies for, and means of coping with, depression between two groups of currently nondepressed adults: one with a history of major depression and one with no history of depression. Currently nondepressed participants were selected so that effects of past experience of depression could be distinguished from those of current mood. Recovered depressed participants (RD) (n = 25) and Never depressed participants (ND) (n = 25) recruited via newspaper advertisements completed self-report measures of (a) the perceived utility of either professional or self-help coping strategies for managing their own experiences of depression; (b) likely effectiveness of several major therapies for depression; and (c) perceived accuracy of several etiological theories of depression. RD participants rated depression as being less amenable to everyday selfhelp methods of coping and more in need of professional intervention. However, RD and ND subgroups did not differ significantly in their perceptions of the plausibility of etiological theories of depression in general, nor in their ratings of the likely helpfulness of major therapies. These data were presented at the annual convention of the Association for Advancement of Behavior Therapy, Miami Beach, November The research was supported by a grant from the Institute for Rational Emotive Therapy. This manuscript is based on a master's thesis completed by the first author. We are grateful to thesis committee members Jim Gray and Michele Carter for comments on earlier versions of this material. Address correspondence to David A. F. Haaga, Department of Psychology, Asbury Building, American University, Washington DC ; dhaaga@american.edu Human Sciences Press, Inc.

2 238 Journal of Rational-Emotive & Cognitive-Behavior Therapy Understanding laypeople's perceptions of the etiology and treatment of depression is important for several reasons. First, beliefs about the optimal treatment of depression may affect treatment acceptability and even treatment outcome. For example, depressed patients appear to respond better to cognitive therapy of depression if they have an initial favorable reaction to the theoretical rationale for this treatment (Fennell & Teasdale, 1987), which may in turn depend upon the goodness of fit of this rationale with their own understanding of how depression is caused (Addis & Jacobson, 1996). Second, lay beliefs about psychological disorders may be a factor in the social context of the disorder (e.g., Furnham & Haraldsen, 1998). If, for example, friends, relatives, co-workers and acquaintances tend to perceive depression as similar to ordinary sadness and as easily overcome via straightforward coping tactics such as distracting oneself by watching TV, this may shed light on depressed people's experiences of others as often impatient, lacking in empathy, and even hostile and rejecting (e.g., Feldman & Gotlib, 1993). Several studies have examined lay perceptions of depression. Research by Rippere (e.g., 1977, 1979, 1980) established that lay people on average have a detailed and reasonably accurate set of beliefs about depression and about useful techniques for coping with depression. Nevertheless, there are individual differences in perceptions of depression. Having suffered a depressive episode oneself appears to be one correlate of these individual differences. For example, in one study depressed patients were more likely than other non-professional participants to cite biological or medical causes when asked open-ended questions about what causes depression and were more likely to consider antidepressant medication a useful treatment (Kuyken, Brewin, Power, & Furnham, 1992). It is not known whether such differences would persist after remission of a depressive episode. People who have experienced major depression but are not currently in a depressive state may have a unique vantage point on depression and unique insights (Coyne, 1994). Detailed first-person reports by such individuals are available (e.g., Styron, 1990), but not systematic comparisons with demographically similar never-depressed groups on the basis of standardized measures. Accordingly, in the study reported in this article we contrasted perceptions of theories, therapies, and coping tactics for depression among (a) a community sample of people who had recovered (for at least two months) from a major depressive episode and (b) a demographically similar group of people with no history of major depression.

3 Lindsey Kirk, Cindy Brody, Ari Solomon, and David A. F. Haaga 239 Participants METHOD The participants were 50 adults (25 Recovered Depressed [RD] and 25 Never Depressed [ND] who had responded to newspaper advertisements. To be included in either group, participants needed to meet the following criteria: (a) Currently non depressed (Beck Depression Inventory of 9 or lower, as recommended by Kendall, Hollon, Beck, Hammen, & Ingram, 1987); (b) No history of manic or hypomanic episodes; (c) No history of primary psychotic ideation, and no bizarre behavior of impaired mental status evident on the day of the study; (d) No current suicidality; (e) No substance abuse or dependence (other than nicotine dependence) in the past six months; (f) No use of antidepressant medication or of psychotherapy for depression in the past two months; and (g) Age at least 18 years. ND participants had to meet two additional criteria: (a) No past major depressive episodes (MDE); and (b) No history of dysthymic disorder. RD participants had to meet three additional criteria: (a) Positive history of major depressive disorder by DSM-IV (American Psychiatric Association, 1994) criteria; (b) Experience of at lest one MDE in the previous three years not precipitated and sustained by drug or organic factors; and (c) Complete recovery from the most recent MDE (asymptomatic by SCID criteria) at least two months prior to the study. Requiring at least two months of asymptomatic functioning is consistent with the recommendation of the MacArthur Foundation Research Network on the Psychobiology of Depression (Frank et al., 1991) that at least 8 symptom-free weeks by considered a consensus definition of full recovery from a major depressive episode. Measures Beck Depression Inventory. The Beck Depression Inventory (BDI; Beck, Rush, Shaw, & Emery, 1979) is a self-report measure of current depressive symptom severity, with extensive evidence of high internal consistency and high convergent validity with independent interviewer ratings (Beck, Steer, & Garbin, 1988). As indicated earlier, participants in either group had to score 9 or below on the BDI to be included in the study.

4 240 Journal of Rational-Emotive & Cognitive-Behavior Therapy Inventory to Diagnose Depression Lifetime. The lifetime version of the Inventory to Diagnose Depression (IDD-L; Zimmerman & Coryell, 1987), a 22-item self-report measure designed to correspond to DSM- III depression criteria, was used for sample description purposes and to corroborate the RD/ND diagnoses we derived from structured interviews. Beck Scale for Suicide Ideation. The Beck Scale for Suicide Ideation (BSI; Beck, Steer, & Ranieri, 1988) is a 19-item self-report measure of suicidality that is highly correlated (>.90) with clinical ratings of suicidal ideation (Beck, Steer, & Ranieri, 1988). The first five questions of this measure, which serve as an overview of suicidality, were used to screen out potentially suicidal people (i.e., anyone scoring above 0). Structured Clinical Interview for DSM-IV. Diagnoses were derived on the basis of relevant portions of the non-patient edition of the Structured Clinical Interview for DSM-IV (SCID-I/NP; First, Gibbon, Spitzer, & Williams, 1995). Williams et al. (1992) reported adequate interrater reliability for major depression diagnoses based on an earlier version of the SCID. In the present study SCID interviews were conducted by clinical psychology Ph.D. students, under the supervision of a licensed psychologist with experience using SCID protocols in research and the immediate direction of an advanced Ph.D. candidate with independent SCID-I training and certification and experience conducting SCIDs in several clinical trials. All SCID interviews were audiotaped to facilitate evaluation of interrater agreement. A randomly-selected subset of 20 SCID audiotapes (40% of the total sample, n = 11 RD, n = 9 ND) were submitted to the supervising psychologist for an independent diagnostic evaluation conducted without awareness of the original diagnosis. There was 100% agreement between these judgements. Helpfulness of Antidepressive Activities Questionnaire. The Helpfulness of Antidepressive Activities Questionnaire (Rippere, 1979) consists of a list of 15 things "that people sometimes do when they are feeling depressed or low." Respondents give a rating from zero (least helpful) to 20 (most helpful), based on how helpful they think they would find each activity. Of the 15 items, we identified 13 a priori as reflecting genuinely common sense everyday self-help activities (e.g., "go for a walk," "watch TV," "see a friend"). Two other items ("read about depression and how to cure it" and "take antidepressants") ap-

5 Lindsey Kirk, Cindy Brody, Ari Solomon, and David A. F. Haaga 241 peared to reflect instead a somewhat more professional approach to managing depression. Accordingly, we conducted analyses involving this measure with the 13-item and 2-item subsets separately; in each case, results are reported in terms of the mean item score on the subscale. Ratings of Causal Theories and Therapies of Depression. Participants' perceptions of several major etiological theories and therapies of depression were measured via rating scales developed by Kuyken et al. (1992). The rating scales consist first of descriptions of biological, diathesis-stress, cognitive, and psychodynamic theories of depression, each of which the participant rates on seven-point scales (anchored with: "explains the causes of depression" not at all and completely). Next, depression treatments (medication, ECT, psychoanalysis, cognitive therapy) are described, and the participant rates each of these on a seven-point scale as well, based on the perceived helpfulness of the treatment (anchored with not at all helpful and extremely helpful). The descriptions of theories and therapies are approximately words long. Kuyken et al. (1992) wrote the descriptions and edited them on the basis of comments made by clinical psychologists asked to evaluate their accuracy. Semi-Structured Interview on Depression Experience. Participants were asked about any previous academic or professional experience they may have had dealing with depression. In addition, they were asked to describe what types of formal treatment, if any, they had received for their depression. Procedure Participants took part in the study individually. After completion of informed consent, the Beck Depression Inventory (BDI) and The Beck Scale for Suicide Ideation (BSI) were administered. Those scoring 9 or under on the BDI and zero on the BSI (thus remaining eligible for the study) next completed the exclusionary portions of the SCID. Participants who were not excluded based on the SCID then completed the standardized measures regarding lay perceptions of depression, as well as other measures not relevant to this report. Finally, participants were interviewed regarding any past academic or professional experiences with depression, treatment history, and detailed histories of mood disturbances among the RD participants. At the end of this

6 242 Journal of Rational-Emotive & Cognitive-Behavior Therapy interview and testing session, participants were debriefed and paid 20 dollars for their time. Descriptive Data RESULTS Demographics and Depressive Symptoms. Descriptive data on demographics and depression variables are summarized in Table 1. As intended, RD and ND groups were very similar demographically. Corroborating the SCID results used in group classification, the RD group reported much more severe depression in the past on the IDD-L (M = 41.48) than did the ND participants (M = 12.67). RD participants also obtained significantly higher scores on the BDI (M = 3.88) than did the ND group (M = 2.04). Although this is not ideal from the standpoint of clearly distinguishing effects of current vs. past depression, the RD average is still well within the nondepressed range. Table 1 Demographics and Depression Data for Recovered Depressed and Never Depressed Participants Demographics Age % Female % Caucasian Depression History IDD-L # Prior MDEs Current Depressive Symptoms (BDI) RD 37.28(12.18) (12.00) Median = (2.65) ND (13.08) (10.55) 2.04 (2.44) t(48) P.44 < Note. Except as noted otherwise, numbers are means, with standard deviations in parentheses. IDD-L = Inventory to Diagnose Depression, Lifetime version. MDEs = Major Depressive Episodes. BDI = Beck Depression Inventory, n = 25 in each group. RD = Recovered Depressed. ND = Never Depressed

7 Lindsey Kirk, Cindy Brody, Ari Solomon, and David A. F. Haaga 243 Personal Experience. Most participants had little relevant academic or professional experience. Five participants (10% of total sample; 4 RD people, 1 ND) had master's degrees in health-related fields (e.g., nursing, vocational rehabilitation), but none of these was a depression specialist. Thus, the sample appears representative in terms of having largely a lay rather than professional/scholarly perspective on depression. Tests of Primary Research Questions There were no significant sex differences on any of the dependent measures. Also, the distribution of each variable was explored and found not to violate the assumptions of normality. Accordingly, RD/ND comparisons were based on -tests, and male and female participants were considered together in these analyses. Theories and Therapies. Ratings of the explanatory power of biological, diathesis-stress, cognitive, and psychodynamic theories of depression, as well as antidepressant medication, electroconvulsive therapy, psychoanalysis, and cognitive therapy are presented separately for RD and ND subsamples in Table 2. There were no significant differences between RD and ND groups on ratings of any of the theories or therapies. Given that there were no significant differences between the RD and ND groups' ratings of the theories and therapies, all participants were combined for supplementary analyses of which theories and therapies were considered most plausible by the sample as a whole. The biological theory (M = 4.90, SD = 1.15) was significantly more strongly endorsed than either cognitive (M = 4.26, SD = 1.68), t(49) = 2.39, p <.03 or psychodynamic (M = 4.00, SD = 1.67), t49) = 3.21, p <.01, theory. In addition, the diathesis-stress theory (M = 4.59, SD = 1.20) was rated significantly higher than the psychodynamic theory (M = 4.00, SD = 1.67), t(49) = 2.32, p <.03. With respect to therapies, in the sample as a whole cognitive therapy (M = 5.12, SD = 1.43) was endorsed significantly more strongly than either ECT (M = 3.04, SD = 1.43), t(49) = 6.84, p <.001 or psychoanalysis (M = 4.38, SD = 1.75), t(49) = 2.78, p <.01. Antidepressant medication (M = 5.21, SD = 1.53) was also rated significantly higher than either ECT, t(49) = 8.56, p <.001, or psychoanalysis, t(49) = 2.59, p <.02. Finally, psychoanalysis was endorsed significantly more than was ECT, t(49) = -4.59,p <.001.

8 244 Journal of Rational-Emotive & Cognitive-Behavior Therapy Table 2 Ratings of Theories and Therapies of Depression, by Recovered Depressed and Never Depressed Participants Theory Biological Diathesis-Stress Cognitive Psychoanalytic Therapy Drug Electroconvulsive Psychoanalysis Cognitive RD M(SD) 4.68 (1.38) 4.74(1.11) 4.08 (1.73) 4.40 (1.58) 5.02 (1.74) 2.79 (1.59) 4.44 (1.66) 5.04 (1.58) ND M(SD) 5.12 (0.83) 4.44 (1.29) 4.44 (1.64) 3.60 (1.68) 5.40 (1.29) 3.28 (1.24) 4.32 (1.87) 5.20 (1.29) t(df) 1.37 (39.5) -0.88(48) 0.76 (48) (48) 0.88 (48) 1.20 (43.6) -0.24(48) 0.39 (48) P Note, n = 25 in each group. RD = Recovered Depressed. ND = Never Depressed. Meats are reported with corrected degrees of freedom when variances were unequal. Coping Tactics. The RD subgroup endorsed the 13 self-help activities in terms of helpfulness in coping with depression significantly less (M = 11.12, SD = 2.98) than did their ND counterparts (M = 13.18, SD = 2.60), *(48) = 2.59, p <.02, Cohen's (1988) d =.74. Conversely, recovered depressed people gave significantly higher ratings to the two professional antidepressive activities (reading about depression and taking antidepressants) (M = 8.52, SD = 5.03) than did the never depressed participants (M = 5.62, SD = 4.17), t(48) = -2.22, p <.04, d =.63. As indicated earlier, the RD group scored significantly higher on the BDI as a measure of current depressive symptoms than did the ND group. Because of this unintended group difference, we repeated the significant RD/ND analyses on perceptions of coping tactics with statistical control of BDI. An ANCOVA with BDI as covariate again showed the ND group to score significantly higher on the 13 self help items, F (1, 47) = 4.07, p <.05. The higher mean score among RD's than among ND's for the 2 professional items was, however, not significant when we controlled for BDI scores, F(l, 47) = 3.23, p <.08.

9 Lindsey Kirk, Cindy Brody, Ari Solomon, and David A. F. Haaga 245 DISCUSSION Two groups of currently non depressed people one with a history of major depression and one without responded to questions about the utility of various means of coping with depression, and the adequacy of several major explanatory theories and professional therapies for depression. The main difference between the groups' perceptions was that the recovered depressed were especially likely to indicate that they would find the more professional coping strategies (reading about depression, taking medication) helpful, whereas the never-depressed group significantly exceeded the recovered-depressed participants in perceiving self-help strategies as likely to be effective. These differences seem logical in that responses to the Helpfulness of Antidepressive Activities Questionnaire are referenced to the extent to which one perceives the various coping methods as likely to be helpful for one's own depression. Given the much more severe depression the RD subsample had experienced (reflected in IDD-L group differences), it stands to reason that they would be less optimistic about the sufficiency of going for a walk, watching TV, etc. than would the Never-depressed participants. Perceptions of " depression" in the abstract (as opposed to one's own depressions), however, were actually quite similar across groups. There was no difference between the RD and ND groups in their ratings of the adequacy of several theories of the causes of depression, nor in their ratings of the likely utility of some major treatments for depression. Our findings differed from those of the Kuyken et al. (1992) study of currently depressed patients, in that the recovered depressed participants in the present sample did not endorse drug treatment more than did never-depressed people. One possible explanation for the discrepant findings is that depressed people's beliefs about etiology and treatment change after recovery. Another possible explanation is that sampling from an inpatient setting was at least in part responsible for Kuyken et al.'s finding of a preference for biological theorizing and drug treatments among depressed participants. Future research sampling from multiple settings and including both currently and formerly depressed participants could resolve this issue. Interpretation of our null results regarding etiological theories and therapies must be tentative in view of the modest sample size in this project, limiting statistical power for detecting between-group differences. For two-tailed f-tests, with comparisonwise alpha level of.05, and 25 participants in each group, power to detect conventionally-

10 246 Journal of Rational-Emotive & Cognitive-Behavior Therapy defined (Cohen, 1988) "small" effects (d =.20) was just.11, and for "medium" effects.43. Our power was adequate (.81) for detecting a large effect (d =.80). Low power is thus a limitation of the study. However, it should be noted that group mean differences in ratings of theories and therapies were small, not just statistically nonsignificant (see Table 2). Also, our subgroups were well-defined, eliminating some possible sources of nuisance variation and bolstering power even at a small sample size (Hallahan & Rosenthal, 1996). For example, RD and ND groups were matched on sex and ethnicity, and diagnoses were in perfect agreement when evaluated by a second rater. An additional methodological issue is that the tests used to measure our main variables were not psychometrically matched, and our results could conceivably have stemmed from differential reliability rather than differences in the constructs being measured. It is noteworthy in this regard that the test showing significant RD/ND differences (Helpfulness of Anti-Depressive Activities Questionnaire) involved multi-item subscales, whereas the tests not differentiating subgroups (ratings of etiological theories and therapies) consisted of 1-item subscales. Nevertheless, these 1-item subscales did yield significant differences across items, even if not RD/ND differences. Considering the sample as a whole, a biological theory of depression received the strongest endorsement, significantly higher than cognitive or psychodynamic. Both cognitive therapy and antidepressant medication treatments were rated significantly more favorably than psychoanalysis, which in turn was seen as significantly more likely to be helpful than ECT. It might seem that these varying ratings of therapies could simply reflect differential exposure to the therapies in one's treatment history. Given that our sampling procedure selected for patients' having recovered from any previous depressive episodes, their preferring a particular therapy may amount to a testimonial from a satisfied customer rather than a more detached, dispassionate consideration of all the possible treatments. This interpretation is difficult to rule out altogether; for instance, the lowest-rated therapy (ECT) is one that none of our participants had undergone. However, two considerations argue against interpreting the therapy ratings as direct reflections of treatment experience. First, as noted above, the ND group whose ratings are presumably not driven by personal experience with therapy for depression did not differ significantly from the RD group in ratings of any of the treatments and rank-ordered them similarly (see Table 2). Second, to explore this issue

11 Lindsey Kirk, Cindy Brody, Ari Solomon, and David A. F. Haaga 247 further we examined therapy ratings in relation to RD participants' responses to our open-ended questions about past treatments. Of the 25 RDs, 84%, (n - 21) reported having received some sort of treatment for their depression. Of these 21, 15 had taken antidepressant medication, while 8 had received cognitive therapy (with respect to all psychotherapies, the labels are ours based on participants' descriptions, not the participants'), 9 psychodynamic therapies, and 6 described another psychotherapy approach such as gestalt or eclectic (numbers total more than 21 because some participants reported multiple therapy experiences). Among RD participants, ratings of the likely helpfulness of antidepressant medication, psychoanalysis, and cognitive therapy did not differ significantly between those who had experienced that particular treatment and those who had not. Thus, people did not seem to be giving ratings based on which treatments they had personally experienced. ECT is a unique case in that not even the most ardent advocate of this treatment would claim that it is a first-line treatment of choice for most depressed people; rather, it would typically be considered only in severe cases after other treatments had failed. For dynamic therapies, though, the issue arises as to whether one is to some extent fighting an uphill battle against laypeople's perceptions that the therapy is less plausible as a treatment for depression. One issue that should be considered in future research on lay perceptions as a factor in treatment process and outcome is whether optimal results are obtained by matching patients with treatments they already believe in (cf. Addis & Jacobson, 1996) or by therapists taking pains to persuade patients of what might seem initially unlikely approaches by providing a very convincing rationale. REFERENCES Addis, M. E., & Jacobson, N. S. (1996). Reasons for depression and the process and outcome of cognitive-behavioral psychotherapies. Journal of Consulting and Clinical Psychology, 64, American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders, Fourth Edition. Washington, DC: American Psychiatric Association. Beck, A. T., Rush, A. J., Shaw, B. P., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press. Beck, A. T., Steer, R. A., & Garbin, M. C. (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review,

12 248 Journal of Rational-Emotive & Cognitive-Behavior Therapy Beck, A. T, Steer, R. A., & Raineri, W. F. (1988). Scale for suicide ideation: Psychometric properties of a self-report version. Journal of Clinical Psychology, Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd edn.). Hillsdale, NJ: Lawrence Erlbaum. Coyne, J. C. (1994). Self-reported distress: Analog or ersatz depression? Psychological Bulletin, 116, Feldman, L. A., & Gotlib, I. H. (1993). Social dysfunction. In C. G. Costello (Ed.), Symptoms of depression (pp ). New York: John Wiley & Sons, Inc. Fennell, M. J. V., & Teasdale, J. D. (1987). Cognitive therapy for depression: Individual differences and the process of change. Cognitive Therapy and Research, 11, First, M. B., Gibbon, M. G., Spitzer, R. L., & Williams, J. B. W. User's Guide for the Structured Clinical Interview for DSM-IVAxis-I Disorders, (SCID- I, Version 2.0, October 1995 Final Version). Frank, E., Prien, R. F, Jarrett, R. B., Keller, M. B., Kupfer, D. J., Lavori, P. W., Rush, A. J., & Weissman, M. M. (1991). Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Archives of General Psychiatry, 48, Furnham, A., & Haraldsen, E. (1998). Lay theories of etiology and 'cure' for four types of paraphilia: Fetishism; pedophilia; sexual sadism; and voyeurism. Journal of Clinical Psychology, 54, Hallahan, M., & Rosenthal, R. (1996). Statistical power: Concepts, procedures, and applications. Behavior Research and Therapy, 34, Kendall, P. C., Hollon, S. D., Beck, A. T., Hammen, C. L., & Ingram, R. E. (1987). Issues and recommendations regarding use of the Beck Depression Inventory. Cognitive Therapy and Research, 11, Kuyken, W, Brewin, C. R., Power, M. J., & Furnham, A. (1992). Causal beliefs about depression in depressed patients, clinical psychologists and lay persons. British Journal of Medical Psychology, 65, Rippere, V. (1977). Some cognitive dimensions of antidepressive behaviour. Behavior Research and Therapy, 15, Rippere, V. (1979). Scaling the helpfulness of antidepressive activities. Behavior Research and Therapy, 17, Rippere, V. (1980). What makes depressed people feel worse? Behaviour Research and Therapy, 18, Styron, W. (1990). Darkness visible. New York: Random House. Teasdale, J. D. (1985). Psychological treatments for depression: How do they work? Behaviour Research and Therapy, 23, Williams, J. B., Gibbon, M., First, M. B., Spitzer, R. C., Davies, M., Borus, J., Howes, M. J., Kane, J., Pope, H. G., Rounsaville, B., & Wittchen, H. U. (1992). The Structured Clinical Interview for DSM-III-R (SCID): II. Multisite test-retest reliability. Archives of General Psychiatry, 49, Zimmerman, M., & Coryell, W. (1987). The Inventory to Diagnose Depression (IDD): A self-report scale to diagnose major depressive disorder. Journal of Consulting and Clinical Psychology, 55,